© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6665
    Prev
    Next
    INDEX
    Explanations

    critical importance/necessity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åŃIJéĩĮ
    -0.09
    ç´¯äºĨ
    -0.09
    mart
    -0.09
    èī®
    -0.08
    essim
    -0.08
    ores
    -0.08
    çŁ¥ä¹İ
    -0.08
    еÑĨ
    -0.08
    اء
    -0.08
    oooooooo
    -0.08
    POSITIVE LOGITS
    ity
    0.16
    ä¸ĩåĪĨ
    0.12
    -minded
    0.12
    shaw
    0.12
    ITY
    0.12
     jun
    0.12
    ness
    0.12
    æĢ§çļĦ
    0.11
    ité
    0.11
    mente
    0.11
    Activations Density 0.017%

    No Known Activations